Unverified50% confidenceFactExact time
DeepSeek's first-generation model consumed 389,000 bytes per context token, whereas the new model released this month consumes only 890 bytes per token, a 437-fold reduction.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/15/2026
First Seen
Valid until: 12/14/2026
Sources
Related Entities
Related Claims
UnverifiedDeepSeek-Coder具备16K甚至更长的上下文窗口75% similarUnverifiedDeepSeek released a new model, Waze Pro, with 1.6 trillion parameters, 49 billion activated parameters, and support for a 1 million token context window.71% similarUnverifiedThe DeepSeek series models have nearly 13 million downloads on Ollama.71% similarUnverifiedThe latest model released by DeepSeek has pushed the cache read price for context tokens down to $0.003 per million tokens.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/923863API
curl https://kongchang.com/api/v1/knowledge/claims/923863MCP
get_claim(id=923863)